Anthropic Opens Claude Mythos 5 to Enterprises via Claude Security Beta

On August 22, Anthropic announced that Claude Security is now powered by its strongest safety model, Claude Mythos 5, and has entered public beta for all Claude Enterprise customers. This marks the first time Mythos 5 is accessible beyond the small-scale Project Glasswing, letting enterprises apply the model to code scanning without applying for separate model access.

Confirmed

• The security scanning feature runs on Claude Mythos 5 and is in public beta for all Claude Enterprise customers.
• Core capability: point it at a GitHub repository to scan and trace vulnerabilities across files.
• Fix workflow: suggested patches open in the web version of Claude Code, reusing the team's existing model configuration.
• Architecture: defenders can use the scanning capability without direct access to the Mythos model.
• Billing: Claude security patches integrate into Claude Code and are billed at standard token rates.

Why it matters

• This is Mythos 5's first step beyond Project Glasswing's closed testing; although only indirectly exposed via the scanning feature, it signals the model's entry into real enterprise deployment.
• Integrating security scanning into Claude Code and GitHub workflows at standard token pricing lowers the barrier for AI-powered security detection and fills a gap in Anthropic's developer security toolchain.

2026-08-22 ~ 2026-08-22 · 6 related posts

• [source] Claude Security Now Powered by Mythos 5 for Cross-File Vulnerability Scanning — claudeai · 2026-08-22
• [source] Claude Security Patches Integrate with Claude Code, Billed as Standard Tokens — claudeai · 2026-08-22
• Anthropic launches enterprise code security scans on Mythos 5 — AccBalanced · 2026-08-22
• Anthropic opens Mythos 5 access for Claude Enterprise security beta — AccBalanced · 2026-08-22

2 near-duplicate retellings: HankYeomans · rohanpaul_ai